Back to All Shipped SystemsOpen Source Production Codebase
ai_rag
Timeline: 2024 – 2025

Nirapod Probaho (AegisFlow)

AI-orchestrated multi-agent disaster resilience and relief distribution platform.

Architectural Role: Full-Stack & Multi-Agent Lead
Nirapod Probaho (AegisFlow) Architecture Diagram
Verifiable Metrics & Performance Receipts
Agent Crew:4 Multi-Agent Roles
Spatial Engine:PostGIS Spatial
Map Rendering:MapLibre GL 60fps
Technology Stack & Libraries
Next.jsCrewAIGemini 1.5 ProMapLibre GLPostGISFastAPI

1. Problem Context & Objectives

During catastrophic flood events in Bangladesh, relief distribution is hindered by delayed field intelligence, uncoordinated volunteer dispatch, and fragmented logistical communication.

2. Technical Architecture & Implementation

Architected an autonomous multi-agent triage system with CrewAI and Gemini 1.5 Pro executing spatial routing over PostGIS geometries, combined with 60fps vector tile rendering in MapLibre GL.

3. Outcomes, Benchmarks & Empirical Validation

Demonstrated rapid shelter routing, resource bottleneck detection, and automated volunteer dispatch coordination across active flood zones.

Associated Research Papers & Benchmarks (Preprints)

Source Code Inspection & Local Replication

This system is maintained in an open-source repository with full git commit history, environment configuration templates (.env.example), and dependency manifests.